Can a Team’s EQ Be Inferred From Individual Scores?

What does a team average actually tell you?

Averaging is straightforward arithmetic: add the scores from each responding member and divide by the number of responses. The meaning is less straightforward. If everyone answered questions about their own recent emotional habits, the result describes the group’s average self-report on that instrument. It does not directly describe how the group handles tension, notices a colleague’s frustration, or repairs a strained conversation.

The distinction matters because the questions may point at different targets. “I notice when my mood changes” asks about an individual’s self-view. “In this team, we notice when someone is upset” asks members to judge a shared practice. In the latter case, the wording refers to the team; agreement among members helps establish that a team summary is meaningful. The individual question can still be useful, but averaging it does not change its subject from “I” to “we.”

A study that developed the Team-Trait Meta Mood Scale (T-TMMS) made that shift explicit. Its team-referent questions ask whether people in the team attend to one another’s feelings, understand emotions in the group, and try to repair negative moods. The study’s sample count is inconsistent within the accessible article: the abstract reports 1,746 participants in 152 leadership teams, while the Methods section reports 1,764 leaders in 152 teams. The measure’s team-referent focus is the useful point here; neither count changes the distinction between a team-referent measure and an average of individual EQ scores. The T-TMMS also relies on members’ reports, so it records perceptions of team practice rather than direct observation.

Sources: Team Emotional Intelligence in Working Contexts: Development and Validation of the Team-Trait Meta Mood Scale (T-TMMS)

What evidence makes averaging defensible?

A plausible team mean needs more than a calculator. The same defined measure should be used across members, enough of the team should respond to make the summary representative, and the research question should justify combining those particular individual results. When the claim is that members share a team experience, researchers also examine whether responses within each team are sufficiently alike and whether scores reliably distinguish one team from another.

Two common statistics help answer those latter questions. Within-team agreement asks whether members of a team give similar ratings. The intraclass correlation (ICC) describes how much responses cluster by team; a second form of ICC estimates how reliably team means differ. These statistics do different jobs, and none turns a weakly chosen instrument into a good measure of team EQ. Cutoffs are conventions that need interpretation in light of the measure, sample, and purpose.

A separate Frontiers study examined 1,566 managers grouped into 188 teams across 90 firms, alongside 4,564 workers. It measured team EI with the team-referent T-TMMS. The authors checked within-team agreement (rwg) and ICC1/ICC2 for aggregated passion and cohesion; those checks supported aggregation of those two constructs. In the multilevel model, team EI was positively associated with passion, and the authors reported an indirect relationship between transformational leadership and group cohesion through team EI and passion. These results concern a team-referent measure in one cooperative corporation in Spain, not proof that an average individual EQ score captures shared practice. The study relied on self-report measures, with potential common-method and social-desirability bias. The authors note that they collected no longitudinal or qualitative data, so the findings cannot support causal conclusions.

There is also a legitimate case for averaging individual measures when the research question concerns the emotional skills members bring to work that is mostly divided among them. In an exploratory study of 33 self-directed work teams, Frye, Bennett, and Caldwell calculated average scores from individual EQ-i responses and reported associations between averaged interpersonal EI and two team-process measures. That is evidence that a mean can be studied as team composition. It does not show that the mean represents shared norms, and the small, exploratory sample limits how broadly to apply the findings.

The right label therefore depends on both the item wording and the intended claim: ‘average member self-report’ is narrower than ‘team emotional intelligence.’ If the group’s collective response is the question, a team-referent measure and evidence of member agreement are more aligned with it.

Sources: Team Emotional Intelligence in Working Contexts: Development and Validation of the Team-Trait Meta Mood Scale (T-TMMS); Team Emotional Intelligence: Emotional Processes as a Link Between Managers and Workers; Team Emotional Intelligence and Team Interpersonal Process Effectiveness

What should you do with a team score?

Look at the work before treating the mean as a practical verdict. Where members complete separate tasks with few dependencies, a summary of individual skills may help describe the team’s composition. Where decisions, handoffs, and problem-solving depend on colleagues responding to one another, a mean says little about how the interaction unfolds. Team research has distinguished individual-referent and team-referent approaches partly because task interdependence changes which level of emotional skill is relevant.

That does not make a team-referent questionnaire an objective record. Members can disagree, overlook behavior, or rate the same exchange differently. Use its result as a prompt to identify what the team experiences, then compare that account with observable work. If people give sharply different answers about whether feedback is discussed respectfully, the disagreement itself is useful information; a single average would hide it.

Choose one recent handoff, meeting, or disagreement and ask: What feeling or concern went unspoken? What response helped, and what repair would make the next exchange easier? The team can choose one behavior to try, such as checking how a decision landed before moving on. This conversation can test whether a score fits members’ experience without treating it as a cause of the problem or a guarantee of better performance. The evidence reviewed comes from particular measures and settings, including cooperative organizations in Spain and a small exploratory work-team study; other workplaces may need different questions.

Sources: Team Emotional Intelligence in Working Contexts: Development and Validation of the Team-Trait Meta Mood Scale (T-TMMS); Team Emotional Intelligence: Emotional Processes as a Link Between Managers and Workers

Questions readers ask